Global chassis controller (GCC) design for autonomous vehicles relies on the information of the environmental factors, weather conditions, vehicle dynamics, actuation bandwidth, among others. Typically, various sensors and actuators are employed to provide such information. Challenges such as cost of sensors, actuator complexity and constraints, fail-safe operations, control authority allocation, and adaptability to a wide range of driving scenarios such as acceleration/ deceleration at set speed, double lane change, and driving on a circular path among others persist for design of such GCC architectures.
Specifically for longitudinal-vertical vehicle controllers tuned to achieve safety and comfort objectives, the performance is significantly affected by the precise knowledge of road conditions i.e., tire friction and road elevation in the presence of nonlinearities such as aerodynamic drag, rolling resistance, spring and damper nonlinearities. For the longitudinal vehicle motion, tire-road friction conditions, aerodynamic forces, engine friction, and rolling nonlinearities critically affect the design of safety controllers such as traction control or active cruise control. Similarly, for vertical vehicle motion control using active suspension, the random road roughness and road defects, spring and damper nonlinearities, hydraulic actuator nonlinearities, and multi-objective design criteria, make design of controller a challenging task.
With that motivation, the use cost effective virtual sensors to detect such external inputs and subsequent output feedback control solutions for the longitudinal-vertical autonomous vehicle motion is proposed in this book. The focus lies on adaptability of designed controllers and estimators to road friction conditions such as road conditions such as asphalt, snow, ice and the road elevation based on various rough roads and road defects.
Table of Contents
1 Introduction
1.1 Vehicle Dynamcis : Estimation and Control
1.2 Motivations
1.3 Contribution of the Thesis
1.4 Organization of the Thesis
2 Literature Review
2.1 Vehicle Dynamics & Modeling
2.1.1 Modeling Aspects
2.1.2 Longitudinal and Vertical Motion
2.1.3 Effect of Nonlinearities
2.1.4 Effect of Exogenous inputs
2.2 Estimation and Control
2.2.1 Tire friction estimation
2.2.2 Longitudinal Motion Control
2.2.3 Vertical Road Elevation Estimation
2.2.4 Vertical Motion Control
2.3 Sliding Mode for Automotive Applications
2.4 Integrated Longitudinal-Vertical Motion: Estimation & Control
3 Longitudinal Motion Estimation
3.1 Overview
3.2 Nonlinear Vehicle Model
3.3 Robust Observer Design
3.3.1 Unknown Input Reconstruction
3.4 Results
3.4.1 Parameter Selection
3.4.2 Case: Decelerating vehicle speed
3.5 Discussion
3.6 Summary
4 Integrated Longitudinal-Vertical Motion Estimation
4.1 Overview
4.2 Vehicle model
4.3 Robust Observer Design
4.3.1 Unknown Input Reconstruction
4.4 Results
4.4.1 Parameter Selection
4.4.2 Case I: No Speed Control
4.4.3 Case II: PI Speed Control
4.4.4 Discussion
4.5 Discussion
4.6 Summary
5 Output Feedback Vertical Motion Control
5.1 Overview
5.2 Vehicle model
5.3 High Gain Observer
5.4 Robust Sliding Mode Controller
5.4.1 Active Control Objectives
5.5 Results and Discussion
5.5.1 Parameter Selection
5.5.2 Case: Road conditions, uncertainties and sensor noise
5.5.3 Discussion
5.6 Summary
6 Integrated Estimation: Rollover Scenario
6.1 Overview
6.2 Vehicle model
6.3 Robust Observer Design
6.3.1 Adaptive FOSM observer for descriptor system
6.3.2 Modified adaptive STA observer for estimation of unknown inputs
6.4 Results
6.4.1 Discussion
6.5 Summary
7 Conclusions and Future Work
7.1 Conclusion
7.2 Directions for Future Work
7.2.1 Integrated Vehicle Control Perspective
7.2.2 Sliding Mode Perspective
Objective & Research Scope
This book explores the design of robust estimators and control systems for autonomous vehicles, specifically focusing on the longitudinal and vertical motion dynamics. The primary goal is to develop an integrated Global Chassis Controller (GCC) that can reliably estimate external exogenous inputs, such as tire-road friction and road elevation, and regulate vehicle motion to ensure passenger comfort and stability under diverse, dynamic road conditions.
- Development of nonlinear vehicle models encompassing longitudinal, vertical, and roll dynamics.
- Design of robust Higher-Order Sliding Mode (HOSM) and High Gain Observers (HGO) for state and exogenous input estimation.
- Implementation of terminal sliding mode control strategies for active suspension systems.
- Performance evaluation under challenging scenarios including varied road surfaces (asphalt, snow, ice) and road defects (potholes, bumps).
- Validation of integrated control strategies for rollover avoidance and stability in maneuvers.
Excerpt from the Book
3.2 Nonlinear Vehicle Model
In this section we discuss the development of a nonlinear quarter car vehicle model for the longitudinal motion analysis. Considering the effect of nonlinear aerodynamic force (2.1) and the rolling resistance (2.2), the longitudinal acceleration and tire dynamics can be given as [2, 61, 109]
where Jw and Jr are the inertias of hub and ring respectively, ww and wr are the angular velocities on the hub and ring sides respectively, θrw is the torsional angle, kt is tire spring stiffness, bt is tire damping coefficient and Tb is the braking torque. A schematic of the torsional tire model is shown in Fig. 3.1. Now we integrate the dynamics of the vehicle longitudinal motion, the torsional tire (3.1) and the LuGre friction model (2.7) to develop a nonlinear state space
Chapter Summaries
1 Introduction: Provides an overview of vehicle dynamics, the motivation for robust control, and outlines the thesis structure regarding the integrated longitudinal-vertical control approach.
2 Literature Review: Surveys existing methodologies for vehicle modeling, friction and road elevation estimation, and reviews sliding mode techniques applied in automotive systems.
3 Longitudinal Motion Estimation: Develops a modified super-twisting observer for a nonlinear vehicle model to estimate tire-road friction and states during longitudinal motion.
4 Integrated Longitudinal-Vertical Motion Estimation: Proposes a novel robust observer to simultaneously estimate tire friction and ground elevation by integrating vertical and longitudinal vehicle dynamics.
5 Output Feedback Vertical Motion Control: Focuses on active suspension control, using a High Gain Observer and terminal sliding mode control to improve ride comfort and road holding.
6 Integrated Estimation: Rollover Scenario: Extends the methodology to a 4-DOF vehicle model to estimate rollover status and exogenous road inputs using an adaptive modified super-twisting approach.
7 Conclusions and Future Work: Summarizes the key research findings and suggests future directions for integrated vehicle control and sliding mode applications.
Keywords
Autonomous Vehicles, Vehicle Dynamics, Sliding Mode Control, Tire Friction Estimation, Road Elevation, Active Suspension, Observer Design, Nonlinear Systems, Rollover Prevention, Global Chassis Control, HOSM Observer, Vehicle Stability, Exogenous Inputs, State Estimation, Automotive Control
Frequently Asked Questions
What is the core focus of this research?
The research focuses on the robust estimation and control of vehicle dynamics, particularly integrating longitudinal and vertical motion to enhance safety and passenger comfort under varying road conditions.
What are the primary challenges addressed?
The book addresses challenges such as the cost-effectiveness of sensors, the complexity of vehicle actuator constraints, and the need for adaptability in diverse driving scenarios like emergency braking or navigating rough, uneven road profiles.
What is the main objective of the proposed control systems?
The primary objective is to develop a global chassis control (GCC) architecture that can simultaneously manage longitudinal and vertical motions through the use of virtual sensors and robust feedback controllers.
Which mathematical techniques are primarily used?
The work heavily utilizes Sliding Mode Control theory, specifically Higher-Order Sliding Mode (HOSM) observers and Terminal Sliding Mode (TSM) controllers, to achieve finite-time convergence and robust performance.
How is the performance of the controllers validated?
The proposed control schemes are validated through extensive numerical simulations, often integrated with the vehicle simulation software CarSim, to test performance across varied road classes, sensor noise, and parameter uncertainties.
What are the key keywords that define the research?
The research is characterized by terms such as Autonomous Vehicles, Sliding Mode Control, Observer Design, Tire Friction Estimation, Active Suspension, and Rollover Prevention.
How does the LuGre model contribute to this work?
The LuGre friction model is employed as a nonlinear, dynamic model to represent tire-road interaction, allowing the observers to accurately reconstruct friction forces as exogenous inputs even during transient maneuvers.
What is the significance of the rollover index discussed in Chapter 6?
The rollover index, derived from the lateral load transfer ratio, is used to quantify the risk of a vehicle overturning during severe maneuvers, serving as a critical metric for the integrated rollover prevention control system.
- Quote paper
- Kalyana Veluvolu (Author), Jagat Jyoti Rath (Author), 2020, Adaptive Vehicle Estimation and Control for Dynamic Road Conditions, Munich, GRIN Verlag, https://www.grin.com/document/953254
